A longitudinal circulating tumor DNA-based model associated with survival in metastatic non-small-cell lung cancer.

One of the great challenges in therapeutic oncology is determining who might achieve survival benefits from a particular therapy. Studies on longitudinal circulating tumor DNA (ctDNA) dynamics for the prediction of survival have generally been small or nonrandomized. We assessed ctDNA across 5 time points in 466 non-small-cell lung cancer (NSCLC) patients from the randomized phase 3 IMpower150 study comparing chemotherapy-immune checkpoint inhibitor (chemo-ICI) combinations and used machine learning to jointly model multiple ctDNA metrics to predict overall survival (OS). ctDNA assessments through cycle 3 day 1 of treatment enabled risk stratification of patients with stable disease (hazard ratio (HR) = 3.2 (2.0-5.3), P < 0.001; median 7.1 versus 22.3 months for high- versus low-intermediate risk) and with partial response (HR = 3.3 (1.7-6.4), P < 0.001; median 8.8 versus 28.6 months). The model also identified high-risk patients in an external validation cohort from the randomized phase 3 OAK study of ICI versus chemo in NSCLC (OS HR = 3.73 (1.83-7.60), P = 0.00012). Simulations of clinical trial scenarios employing our ctDNA model suggested that early ctDNA testing outperforms early radiographic imaging for predicting trial outcomes. Overall, measuring ctDNA dynamics during treatment can improve patient risk stratification and may allow early differentiation between competing therapies during clinical trials.

Clinical analysis of germline copy number variation in DMD using a non-conjugate hierarchical Bayesian model.

BACKGROUND: Detection of copy number variants (CNVs) is an important aspect of clinical testing for several disorders, including Duchenne muscular dystrophy, and is often performed using multiplex ligation-dependent probe amplification (MLPA). However, since many genetic carrier screens depend instead on next-generation sequencing (NGS) for wider discovery of small variants, they often do not include CNV analysis. Moreover, most computational techniques developed to detect CNVs from exome sequencing data are not suitable for carrier screening, as they require matched normals, very large cohorts, or extensive gene panels. METHODS: We present a computational software package, geneCNV ( http://github.com/vkozareva/geneCNV ), which can identify exon-level CNVs using exome sequencing data from only a few genes. The tool relies on a hierarchical parametric model trained on a small cohort of reference samples. RESULTS: Using geneCNV, we accurately inferred heterozygous CNVs in the DMD gene across a cohort of 15 test subjects. These results were validated against MLPA, the current standard for clinical CNV analysis in DMD. We also benchmarked the tool's performance against other computational techniques and found comparable or improved CNV detection in DMD using data from panels ranging from 4,000 genes to as few as 8 genes. CONCLUSIONS: geneCNV allows for the creation of cost-effective screening panels by allowing NGS sequencing approaches to generate results equivalent to bespoke genotyping assays like MLPA. By using a parametric model to detect CNVs, it also fulfills regulatory requirements to define a reference range for a genetic test. It is freely available and can be incorporated into any Illumina sequencing pipeline to create clinical assays for detection of exon duplications and deletions.